Called “a geek squad of civic-minded number-crunchers” by the New York Times in a recent profile, the predictive analytics team led by City of New York’s Chief Analytics Officer Mike Flowers is pioneering new approaches in the field of civic data. Working out of the Mayor’s office — in concert with the efforts of our 2013 NYC Fellows and the local Brigade — this groundbreaking group of civic hackers inside City Hall is pushing forward some of the most innovative and effective uses of civic analytics — and they have the results to show for it.
But they also have time, staff, and executive support that are likely out of reach for other cities. Can other cities implement a civic analytics initiative on a smaller scale and still achieve some of the same successes?
According to Nicholas O’Brien, chief of staff for Mike Flowers’ Office of Policy and Strategic Planning in City of New York, the answer is yes. O’Brien joined this month’s Municipal Innovation discussion to share an inside look at the process their team has used to tackle some of New York’s toughest challenges with big data and predictive analytics — and outlined some strategies for how these data-driven approaches can be applied in other cities, even those lacking the dedicated resources and bandwidth they have in New York.
In New York, quick delivery on early projects helped generate buy-in from other departments and stakeholders. “The goal isn’t to figure out how to run the city perfectly, it’s to improve over what is being done now,” said O’Brien. “Do it quick, test, and iterate” to show small but tangible results initially, and build the support that will allow you to take on bigger initiatives.
Start with open data
“Most of what we use is public data,” said O’Brien. “It’s a great place to start.” Chances are your city already publishes some data online, either through an open data portal or otherwise. Rather than expending resources trying to get other departments to release the data you want, start with what’s already available. Then, the successes of your analytics program can create more momentum for open data within your city — which in turn will give you more material to work with.
Use tools that are readily available
You don’t need to invest in lots of expensive and complicated technology to pilot an analytics initiative. Does your department use Microsoft Excel? That’s a great place to start. “Excel can handle it,” says O’Brien. “There’s a ton of stuff you can use there.”
Build your skillset with free online courses
Nor do you need to hire new skilled staff, or invest in costly training, to get started with basic civic analytics. Organizations like Coursera and EdX offer online courses and training on data science, statistics, and programming that are accessible, high-quality, and most importantly, free. Try R is another free course specifically about learning R, a programming language for statistics and data modeling.
Visualize data through maps
“People respond to maps — they identify with their city, with the location,” said O’Brien. From a marketing and community engagement perspective, maps are an ideal way to get your analytics initiative off the ground and build interest in an a way that’s accessible to a non-technical audience.
Take a page out of NYC’s playbook
One of the first initiatives in New York had to do with matching locations between different datasets, since each agency has a different way to collect and identify data — the buildings department uses the building ID number, the police department uses cartesian coordinates, etc. The locations are relatively easy to match, since they don’t move around and it’s still a single location. Once they had matched locations, the team was able to run analysis that targeted illegally converted apartments — leading to a five-fold return on the time building inspectors spent identifying and inspecting these illegal dwellings.
This location-matching approach is replicable in other cities, says O’Brien: “Every place has locations — there’s maps. You can start to take those ontologies that agencies have developed and bring it into unified format.” From there, the possibilities for what you’ll find in that data are greatly amplified.
“In NYC, we like to think we’re unique and special but the core of how cities operate is similar,” said O’Brien. Many cities in our Peer Network who participated in the Municipal Innovation discussion are gearing up for their own data-driven initiatives — including San Leandro, Calif., Montgomery County, Maryland, and Oakland, Calif. — and we’re looking forward to seeing what happens as the geek squads in City Halls across the county continue to gain momentum.
Has your city experimented with civic analytics? We’d love to hear about it – email us at pn-staff [at] codeforamerica.org and tell us how it worked.
Code for America’s open Municipal Innovation discussions are held bimonthly. Interested in joining the next one in June? Sign up here.
The Municipal Innovation discussions are a sample of the offerings of the Code for America Peer Network, CfA’s new professional learning association for innovators in local government. Learn more here.